A multi-scale approach to understand the mechanobiology of intermediate filaments

Abstract

The animal cell cytoskeleton consists of three interconnected filament systems: actin microfilaments, microtubules and the lesser known intermediate filaments (IFs). All mature IF proteins share a common tripartite domain structure and the ability to assemble into 8–12 nm wide filaments. At the time of their discovery in the 1980s, IFs were only considered as passive elements of the cytoskeleton mainly involved in maintaining the mechanical integrity of tissues. Since then, our knowledge of IFs structure, assembly plan and functions has improved dramatically. Especially, single IFs show a unique combination of extensibility, flexibility and toughness that is a direct consequence of their unique assembly plan. In this review we will first discuss the mechanical design of IFs by combining the experimental data with recent multi-scale modeling results. Then we will discuss how mechanical forces may interact with IFs in vivo both directly and through the activation of other proteins such as kinases.

Introduction

In metazoan cells, the actin microfilaments (MFs) and microtubules (MTs) cytoskeleton is accompanied by a third filament system arising from the cell-type-specific expression of a complex gene family that in man encodes more than 70 proteins. These fibrous proteins are absent from both plants and fungi and harbour a common tripartite organization characterized by a central α-helical coiled-coil-forming domain and non-α-helical “head” and “tail” domains of variable length and sequence (Strelkov et al., 2003). As other fibrous proteins such as collagen (Yuan and Veis, 1973) or fibrin (Cavazza et al., 1981), they self assemble into filaments with a diameter of ∼10 nm (Herrmann et al., 1996). Their name, intermediate filaments (IFs), comes from the fact that their diameter is intermediate between the ones of actin and myosin filaments (Ishikawa et al., 1968).

The cell-specific IF network is often pictured as an integrator of MFs and MTs via a complex set of cross bridging proteins (Seifert et al., 1992) (Fig. 1). Furthermore, in all tissues, IFs form transcellular networks by direct interaction with specific cell–cell and cell–matrix junctions, desmosomes and hemidesmosomes, respectively (Troyanovsky et al., 1993; Mogensen et al., 1998). Hence each specific IF network integrates the cytoskeletal system of each cell into tissues and organs. As far as the key cell functions are concerned, i.e. cell adhesion and migration, organelle shaping and positioning, IFs seem to play the role of a mediator by providing cellular signposts through post-translational modifications (Omary et al., 2006; Pallari and Eriksson, 2006; Hyder et al., 2008). This constitutes the strongest functional difference to both the MT and MF systems, which are the main players in the above-mentioned cellular functions. This may explain why mutations in their subunit proteins, tubulin and actin, are much less tolerated than in IF proteins. Nevertheless, recent work has revealed a multitude of disease mutations in various IF proteins leading to very complex disease entities directly reflecting the complex expression patterns of IF genes (Omary et al., 2004). Among all IF proteins, lamin A is the one with the highest number of identified mutations, 230 so far, causing a complex set of at least 13 different human diseases including lipodystrophies, muscular dystrophies and even progeria syndromes (van der Kooi et al., 2002; Hegele, 2003, Hegele, 2007; Sylvius and Tesson, 2006; Vlcek and Foisner, 2007). Although the disease mechanism is not at all clear in any of these cases, several models involving stress, cell fate and gene expression have been proposed (Cohen et al., 2008). In addition, disturbances in signalling pathways, as caused by mutations in IF proteins, may be responsible for some aspects of IF diseases in general.

Over the last ten years, IFs have indeed emerged as signalling platforms (Pallari and Eriksson, 2006). For example IF proteins such as vimentin, GFAP, K17 and K18 bind to different isoforms of one of the major families of signalling proteins named 14-3-3 (Kim and Coulombe, 2007). In fact there are demonstrated interactions between IF proteins and other components of the translational apparatus (Traub et al., 1998; Lin et al., 2001; Haddad et al., 2002; Kim et al., 2007). Several studies have also demonstrated that IF proteins attenuate the response to specific proapoptotic signals in several cell culture models and physiological settings but promote apoptosis once commitment to execute is made (Kim and Coulombe, 2007). However, one should not forget that prior to these recent discoveries, IFs were only considered as “mechanical integrators of cellular space” (Lazarides, 1980). This original statement was mainly based on the fact that in a muscle cell, desmin IFs are attached to the nuclear surface, to the mitochondria, to the desmosomes, to the sarcomeres at the Z-disc and to the nucleus. In the case of nucleus and mitochondria the protein mediators have only been identified recently, nesprin-3 (Wilhelmsen et al., 2005) and plectin 1b (Winter et al., 2008), respectively. Based on this peculiar intracellular organization and the various signalling functions mentioned above, it seems reasonable to propose a pivotal role of the IF network in mediating mechanotransduction processes. In this review, we will present our current understanding of IF mechanical design, some experimental evidences of IF networks remodeling by external mechanical stresses in vivo and complementary in silico approaches to test the role of IFs in mechanotransduction events.